Association between Long-Term Exposure to Particulate Matter Air Pollution and Mortality in a South Korean National Cohort: Comparison across Different Exposure Assessment Approaches.
Int J Environ Res Public Health. 2017 Sep 23 ;14(10). Epub 2017 Sep 23. PMID: 28946613
Increasing numbers of cohort studies have reported that long-term exposure to ambient particulate matter is associated with mortality. However, there has been little evidence from Asian countries. We aimed to explore the association between long-term exposure to particulate matter with a diameter≤10 µm (PM10) and mortality in South Korea, using a nationwide population-based cohort and an improved exposure assessment (EA) incorporating time-varying concentrations and residential addresses (EA1). We also compared the association across different EA approaches. We used information from 275,337 people who underwent health screening from 2002 to 2006 and who had follow-up data for 12 years in the National Health Insurance Service-National Sample Cohort. Individual exposures were computed as 5-year averages using predicted residential district-specific annual-average PM10 concentrationsfor 2002-2006. We estimated hazard ratios (HRs) of non-accidental and five cause-specific mortalities per 10 µg/m³ increase in PM10 using the Cox proportional hazards model. Then, we compared the association of EA1 with three other approaches based on time-varying concentrations and/or addresses:predictions in each year and addresses at baseline (EA2); predictions at baseline and addresses in each year (EA3); and predictions and addresses at baseline (EA4). We found a marginal association between long-term PM10 and non-accidental mortality. The HRs of five cause-specific mortalities were mostly higher than that of non-accidental mortality, but statistically insignificant. In the comparison between EA approaches, the HRs of EA1 were similar to those of EA2 but higher than EA3 and EA4. Our findings confirmed the association between long-term exposure to PM10 and mortality based on a population-representative cohort in South Korea, and suggested the importance of assessing individual exposure incorporating air pollution changes over time.